Regulatory factor X 7 (Rfx7) is an uncharacterized transcription factor belonging to a family involved in ciliogenesis and immunity. Here, we found that deletion of Rfx7 leads to a decrease in natural killer (NK) cell maintenance and immunity in vivo. Genomic approaches showed that Rfx7 coordinated a transcriptional network controlling cell metabolism. Rfx7–/– NK lymphocytes presented increased size, granularity, proliferation, and energetic state, whereas genetic reduction of mTOR activity mitigated those defects. Notably, Rfx7-deficient NK lymphocytes were rescued by interleukin 15 through engagement of the Janus kinase (Jak) pathway, thus revealing the importance of this signaling for maintenance of such spontaneously activated NK cells. Rfx7 therefore emerges as a novel transcriptional regulator of NK cell homeostasis and metabolic quiescence.
Access optionsAccess options
Rent or Buy article
Get time limited or full article access on ReadCube.
All prices are NET prices.
Subscribe to Journal
Get full journal access for 1 year
only $18.75 per issue
All prices are NET prices.
VAT will be added later in the checkout.
Publisher’s note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
We thank M. Thome-Miazza and P. Schneider (UNIL, Lausanne); W. Reith (UNIGE Medical School, Geneva); M. Rüegg and M. Hall (Biozentrum, Basel); and M. Mazzone (VIB, Leuven) for sharing reagents important for this study. We thank N. Fonta, W. Held, R. Bedel, S. Siegert, S. Calderon, K. Harshman, and R. Hertzano for technical help. Studies in the group of G.G. were funded by the European Research Council (ERC-2012-StG310890) and the Swiss National Science Foundation (PP00P3_139094 and PP00P3_165833). S.P.M.W. is supported by the ETH post-doctoral fellowship program (FEL29 15-2) and the Horten Foundation. The laboratory of E.V. is supported by the ERC under the European Union’s Horizon 2020 research and innovation programme (grant agreement 694502), Agence Nationale de la Recherche, Innate Pharma, MSDAvenir, Ligue Nationale contre le Cancer (Equipe labelisée ‘La Ligue’) and Marseille-Immunopole. O.B. was supported by the Swiss National Science Foundation 310030-172978 and Swiss Cancer Research KFS-4136-02-2017. M.E.R. was supported by KFS-MDPhD-3557-06-2015. M.D. was supported by the Fondation Medic, Lausanne. P.-C.H. was supported by the Swiss National Science Foundation (31003A_163204), the Swiss Cancer League (KFS-3949-08-2016), and a Melanoma Research Alliance Young Investigator award.
Integrated supplementary information
(a) Percent identity matrix of full RFX7 protein sequences generated with Clustal 2.1. (b) Phylogenetic tree based on full protein alignment (Neighbor-joining tree without distance corrections, Clustal 2.1); the bar line (distance scale) indicates the percentage of variation among species, i.e. 0.1 represent 10% difference between two sequences. (c) Alignment of RFX7 DNA Binding Domain for the indicated species (JalView). The compared protein sequences correspond to the following species: Human, Homo sapiens; Chimpanzee, Pan troglodytes; Gorilla, Gorilla gorilla; Macaque, Macaca mulatta; Orangutan, Pongo pygmaeus; Mouse, Mus musculus; Rat, Rattus norvegicus; Rabbit, Oryctolagus cuniculus; Guinea pig, Cavia porcellus; Pig, Sus scrofa; Cow, Bos taurus; Chicken, Gallus gallus; Xenopus, Xenopus tropicalis; Zebrafish, Danio rerio. (d) Scheme showing the targeting strategy for the conditional knockout of the murine Rfx7 gene. The vector was designed such as the 5’ loxP site is inserted upstream of exon 3 and the target region is 1.96 kb including exons 3-4. The loxP/FRT flanked Neo cassette is inserted downstream of exon 4. The selection cassette (Neo) was excised by crossing floxed mice with FLP deleter line. The two loxP sites shown allow Cre-mediated deletion of exons 3 (Ex 3) and 4 (Ex 4). PCR primers b/c discriminate wild type (wt) and floxed (fl) Rfx7 alleles and primers a/c wild type and knockout (ko) Rfx7 alleles. Arrows on diagram indicate PCR primer positions. PCR products obtained for each genotype are shown. LA: long arm, MA: middle arm, SA: short arm, FRT: flippase recognition target, Neo: neomycine cassette
(a,b) Quantitative RT-PCR (qRT-PCR) data (relative to Polr2a) showing Rfx7 transcript abundance in MACS-sorted T and B cells (a), and in FACS-sorted NK lymphocytes (b) from Vav Rfx7fl/fl and Rfx7fl/fl mice. Results represent mean ± SD of technical replicates (n = 3) and are representative of at least two experiments (a,b)
(a) The abundance of Rfx7 mRNA was determined by qRT-PCR in the indicated subsets of sorted NK cells from BM or spleen (SP) of wild type mice (relative to Polr2a). (b) A representative flow cytometric plot of NK cells (gated as NK1.1+CD3–CD19–) in the blood and liver of Ncr Rfx7wt/wt, and Ncr Rfx7fl/fl mice is shown (gated on CD45+ lymphocytes). Results represent the mean ± SD of n = 3 technical replicates (a) or the mean ± SEM of n = 3 mice/group (b) and are representative of at least two independent experiments (a,b). (b) Statistical comparison between the experimental condition lacking Rfx7 and control were performed; *p ≤ 0.05; Student’s t-test
(a) Control and Rfx7-deficient NK cells were isolated from BM (sorted as CD122+ NK1.1+) and spleen (sorted as Ncr1+ and NK1.1+) from a pool of nine Vav Rfx7 wt/wt:Vav Rfx7 fl/fl mixed BM chimeras. Transcript abundance of 12 genes, selected based on the RNA-sequencing results, was determined by qRT-PCR (relative to Polr2a). (b) MACS-sorted CD4+ and CD8+ T cells from spleens of Cd4 Rfx7wt/wt and Cd4 Rfx7fl/fl mice were tested for mRNA levels of the indicated genes by qRT-PCR (relative to Polr2a). (c) Expression of the indicated proteins in MACS-enriched T cells and T cell-depleted fractions (flow through, FT) from splenocytes of Vav Rfx7wt/wt or Vav Rfx7fl/fl mice was determined by immunoblot analysis. Actin was used as loading control. (d,e) Table illustrating the fold difference (KO:WT) detected in the RNA-sequencing for Rfx genes (d) or the indicated MHC-related genes (e) in BM or spleen NK cells. f) Graphs depict the geometric MFI of surface H2-D, H2-K, and Qa2 as detected on NK cells of the indicated genotypes and a representative histogram thereof. Results depict mean ± SD of n = 3 technical replicates per genotype/organ (a) or the mean ± SEM of n = 3 (Cd4 Rfx7wt/wt) and n = 5 (Cd4 Rfx7fl/fl) mice (b) and of n = 3 (Rfx7fl/fl), n = 3 (Vav Rfx7wt/wt) and n = 4 (Vav Rfx7fl/fl) mice (f). Results are representative of at least two independent experiments (a,b,c,f). (a,b,f) Statistical comparison between the experimental condition lacking Rfx7 and controls were performed; *p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001; Student’s t-test
(a) In Silico analysis of promoter sequences (-700; +300) of differentially expressed genes from cluster 1 (downregulated genes), cluster 2 (upregulated genes), and cluster 0 (non-modulated genes, serving as control group) was performed with JASPAR tool for the presence of putative transcription factor binding sites (TFBS) for CCAAT/enhancer binding protein beta (Cebpb) and T-box 15 (Tbx15). The box and whisker plots depict the number of putative TFBS identified with a relative profile score threshold of 0.8. Significance was determined using one-tail Mann-Whitney tests; *p ≤ 0.05
(a) Quantification of forward scatter (FSC) and side scatter (SSC) for splenic NK cells (gated as NK1.1+ CD3/19-) from Ncr Rfx7wt/wt (set as 100%) and Ncr Rfx7fl/fl mice. (b) A representative ImageStream picture of NK cells (bright field or anti-NK1.1 staining) from Ncr Rfx7wt/wt and Ncr Rfx7fl/fl. Histogram illustrates the surface area based on the bright field of 30’000 to 100’000 cells per mouse and genotype and the graph a quantification thereof (each line colored in hues of red or blue represents one mouse, Ncr Rfx7wt/wt and Ncr Rfx7fl/fl, respectively). (c,d) FSC and SSC for BM NK cells (gated as CD122+ CD3/19–) are shown for Vav Rfx7wt/wt, and Vav Rfx7fl/fl mice (c) and for Ncr Rfx7wt/wt and Ncr Rfx7fl/fl mice (d). (e) Quantification of FSC and SSC for splenic CD4+ T cells (gated as CD3+CD4+), CD8+ T cells (gated as CD3+CD8+), NKT cells (gated as NK1.1+CD3+), NK cells (gated as NK1.1+CD3–), B cells (gated as CD19+), and conventional dendritic cells (DC; gated as CD11chiCD11bint–hi) from the indicated mice. (f) HEK293T cells were co-transfected with WT, NLS-mutant (m658 & 674) Rfx7, or empty (mock) and GFP-encoding vectors. After 48 hours, FSC was analyzed on the GFP+ cells, with the FSC of mock-transfected cells set at 100. (g) Quantification of FSC and SSC for splenic NK cells, CD8+ T cells, and B cells from Rfx5+/+, Rfx5+/–, and Rfx5–/– mice. (h) The graph depicts ratios of Rfx7-deficient to control living NK cells (congenically marked) co-cultured in the presence of high IL-15 and S63845 (Mcl-1 inhibitor; Cayman Chemical) for three days (normalized to initial mix). Flow cytometry plots illustrate dead cell percentage. Results represent the mean ± SEM of n = 6 (Ncr Rfx7wt/wt) and n = 7 (Ncr Rfx7fl/fl) (a,d), n = 5 (Ncr Rfx7wt/wt) and n = 8 (Ncr Rfx7fl/fl) (b), n = 5 (Vav Rfx7wt/wt) and n = 6 (Vav Rfx7fl/fl) (c), n = 8 (Rfx7fl/fl), n = 10 (Vav Rfx7wt/wt), and n = 9 (Vav Rfx7fl/fl) (e), n = 4 (Rfx5+/+), n = 7 (Rfx5+/–), and n = 6 (Rfx5–/–) mice (g) or mean ± SD of n = 4-5 technical replicates per condition (f,h) and are representative of at least two independent experiments (a-f) and a pool of two independent experiments (g,h). Statistical comparison between the experimental condition and controls were performed; ***p ≤ 0.001; Student’s t-test (a-g)
Supplementary Figure 7 Rfx7-deleted NK cells present only moderate alterations in functional features.
(a) Expression analysis of the indicated receptors on splenic NK cells (NK1.1+CD3–CD19–) from Vav Rfx7wt/wt, and Vav Rfx7fl/fl mice is depicted as percentage of positive population (for biphasic expression) or geometric MFI. (b) Percentages and numbers of NK cells (gated as NK1.1+ and CD3/19–) in the spleen of Ncr Rfx7fl/fl and Ncr Rfx7wt/wt mice treated with IL-2 complexes (cIL-2) for four days and used for B2m–/– splenocyte rejection on day 5 (presented in Fig. 8c) are depicted. (c) A representative flow cytometric plot and a quantification of IFNγ and granzyme A production by splenic NK cells of Rfx7fl/fl, Vav Rfx7wt/wt, and Vav Rfx7fl/fl mice are shown. Results represent the mean ± SEM of n = 5 (Vav Rfx7wt/wt) and n = 4 (Vav Rfx7fl/fl) (a) or n = 4 (Ncr Rfx7wt/wt) and n = 5 (Ncr Rfx7fl/fl) mice (b), or n = 4 (c) and are representative of at least two independent experiments (a-c). Statistical comparison between the experimental condition lacking Rfx7 and controls were performed; *p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001; Student’s t-test